http://informahealthcare.com/dre ISSN 0963-8288 print/ISSN 1464-5165 online Disabil Rehabil, 2014; 36(23): 1931–1942 ! 2014 Informa UK Ltd. DOI: 10.3109/09638288.2014.882419

REVIEW

Is there a relationship between pain and psychological concerns related to falling in community dwelling older adults? A systematic review Brendon Stubbs1, Elizabeth West1, Sandhi Patchay2, and Pat Schofield1 School of Health and Social Care, University of Greenwich, Eltham, London, UK and 2School of Counselling and Psychology, University of Greenwich, Southwood Site, Eltham, London, UK

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

1

Abstract

Keywords

Purpose: To systematically review and synthesise the research evidence linking pain to psychological concerns about falling in community dwelling older adults. Methods: A systematic review was conducted in accordance with the preferred reporting items of systematic reviews and meta-analysis statement (PRISMA). Major electronic databases were searched from inception until June 2013. Two authors independently conducted the searches, extracted data and completed methodological quality assessments. Articles were included if they measured one of the psychological concerns related to falling in a sample of community dwelling older adults with pain, or explored the association between the two. Results: Of a potential 892 articles, 12 met the eligibility criteria (n ¼ 3398). The methodological quality of the included studies was variable and none of the included studies primary aim was to investigate the relationship between pain and psychological concerns related to falls. Two studies found significant differences in psychological concerns related to falls in older adults with pain and a control group. Nine out of 10 studies reported a significant correlation between pain and psychological concerns related to falls in their sample. Conclusion: This review provides provisional evidence that pain is associated with fear of falling (FOF), avoidance of activities due to FOF and falls efficacy in community dwelling older adults.

Balance confidence, chronic pain, falls efficacy, fear of falling, musculoskeletal pain, older adult, pain History Received 25 July 2013 Revised 10 December 2013 Accepted 8 January 2014 Published online 28 January 2014

ä Implications for Rehabilitation  

 

Pain is a common and pervasive problem in community dwelling older adults and can affect an individual’s mobility, levels of physical activity and increase their falls risk. Psychological concerns related to falls, such as fear of falling (FOF), falls efficacy and balance confidence are also common and troublesome issues in older adults, yet the association with pain has not been investigated with a systematic review. This review provides provisional evidence that pain may increase older adult’s risk of developing FOF, avoiding activities due to a FOF and impact their falls efficacy. In recognition of the findings of this review, clinicians working with older adults with pain should consider assessing psychological concerns related to falls and if necessary intervene if they identify an individual at risk.

Introduction Falls constitute a common problem in old age with over 30% of community dwelling older adults experiencing one or more falls each year and the risk steadily increases with age [1]. This is a concern since falls are a leading cause of accidental death and disability in older adults [1–3]. For these reasons the prevention of falls is an international priority [4]. Although most older adults who fall do not experience a physical injury, many develop psychological concerns related to falling which can be equally disabling and disruptive upon an individual’s activities of daily living, health, and wellbeing [5,6]. Address for correspondence: Brendon Stubbs, School of Health and Social Care, University of Greenwich, Southwood Site, Eltham, London SE9 2UG, UK. E-mail: [email protected]

It has been demonstrated that concerns about falling are not limited to people who have experienced a fall; many older adults are afraid of falling, even though they have not experienced a fall themselves [7]. A range of psychological concerns related to falling have been studied in recent years but the three most significant and common constructs considered are fear of falling (FOF), falls efficacy and balance confidence [5,7,8]. Recent attention has also been given to one further construct; concerns about the consequences of falling (COF). Each of these constructs has been defined in the literature, but there is some inconsistency in the measurement and reporting of each construct (see [5]). FOF refers to a lasting concern about falling that leads to an individual not performing activities they are capable of doing [9]. Falls efficacy is based on Bandura’s concept of self-efficacy [10] and refers to an individual’s assessment of their own self-efficacy to avoid a fall, while balance confidence refers to an individual’s

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

1932

B. Stubbs et al.

confidence that they will be able to maintain their balance and not fall over when doing their activities of daily living [11]. In this article, we refer to these concepts together as ‘‘psychological concerns related to falling’’, but will discuss individual constructs when authors make specific reference to one or more of the constructs. While worrying about falling and avoiding activities may be functional in the short term and promote safety, for some older adults, FOF is disproportionate to their physical capabilities so that they are restricting their activities unnecessarily [8]. Severe activity restriction can, in turn, create a series of problems, including sensorimotor deconditioning and reduced balance which increases the risk of actual falls [6,12,13]. In addition, FOF is in itself an established risk factor for future falls [3,13]. Many community dwelling older adults are concerned about the FOF; indeed current prevalence estimates are that up to 85% of this group are affected [7]. Several studies have investigated factors predisposing older adults to develop psychological concerns related to falls. These include actual falls, increasing age, female gender, dizziness, depression, anxiety and problems with gait and balance [7,8,14]. Understanding and identifying risk factors provides clinicians and researchers with valuable information, helping them identify high risk groups so that appropriate interventions can be offered. However, one potential risk factor that appears to have received minimal consideration is pain. This is surprising since other fear related behaviours, such as the avoidance of physical activity in older adults experiencing pain (although not in relation to FOF) have been the focus of research in recent years [15–17]. Pain is a common and pervasive problem in community dwelling older adults and it is frequently underdiagnosed [18]. Pain can impair an individual’s balance, gait and mobility [19], is associated with reduced levels of physical activity [20] and with actual falls [21]. Although many of these correlates of pain have been associated with psychological concerns related to falling, the direct association with pain itself remains elusive. Since pain and psychological concerns related to falls are commonly experienced by community dwelling older adults, we conducted a systematic review of the literature to find out whether there is evidence of an association between the two. The aims of the review are: (a) to describe the various psychological concerns related to falling that have been measured in older adults with pain, (b) to investigate whether psychological concerns about falling are more common in older adults with pain than in older adults without pain and (c) investigate any associations reported in the literature between pain and psychological concerns related to falling in community dwelling older adults.

Method This systematic review was conducted and reported in accordance with the preferred reporting items for systematic reviews and meta-analysis statement (PRISMA; [22]). Eligibility criteria To be eligible for inclusion in the review, studies had to satisfy the following criteria: (1) conducted in a sample of older adults dwelling in the community with a mean age of 60 years and older. (2) Measured one of the following psychological constructs FOF, falls efficacy, balance confidence, avoidance of activities due to a FOF, concerns about the COF, with a validated multi-item questionnaire (e.g. falls efficacy scale) or a single item question. (3) Assessed pain in the sample. As we expected there to be few studies in the area, we included studies that investigated any type of pain in the review.

Disabil Rehabil, 2014; 36(23): 1931–1942

We excluded studies conducted in patients with dementia or with recent orthopaedic trauma (within the last year) or orthopaedic surgery (e.g. total hip replacement). The design of the study was not limited; if randomised control trials (RCT) or controlled clinical trials (CCT) were found we used the baseline data to ascertain the variables of interest. We included studies that did not have a control group due to the paucity of data in this field [23]. Only primary data papers providing quantitative data were eligible; reviews, case studies, expert opinions pieces and abstracts were excluded. Only articles written in English were considered for the review. Information sources We conducted a systematic review of the literature searching major electronic databases from inception until 1 June 2013 including Cochrane Library, CINAHL, EBSCO, PubMed and PsycINFO. In addition, the reference lists of all eligible articles and recent systematic reviews of the literature were scanned for eligibility. Searches of the online ‘‘in press’’ sections of key journals in the field were also conducted. Systematic search strategy The medical subject headings used were ‘‘fear of falling’’ OR, ‘‘falls efficacy’’, OR, ‘‘fall related psychological concern’’, OR ‘‘balance confidence’’ OR ‘‘consequences of falling’’ AND ‘‘pain’’, OR ‘‘chronic pain’’, OR ‘‘musculoskeletal pain’’ AND ‘‘older adult’’, OR ‘‘old age’’ OR ‘‘elderly’’. Study selection Two reviewers independently conducted the searches following a predetermined search strategy screening article titles, key words and abstracts to assess the eligibility. Articles that appeared to meet the eligibility criteria were included for consideration in the full text review. The final list for the full text review was decided by consensus by two authors. Two authors then independently conducted the full text review and then decided on the final included articles by consensus. A third reviewer was available for mediation at this stage. If we encountered overlapping studies, we included the most recent and/or the study with the largest sample size. If we required any clarification to inform our discussion as to whether an article met the study eligibility criteria we contacted the author up to three times; if we did not receive a response the article was excluded. Data collection Two authors were involved in the data extraction process using a predetermined form. The data collected from each article included: study design, setting, sample characteristics (number, age, % female gender), method of pain assessment, method of assessment of fall related psychological concern (including reporting of psychometric properties if available), prevalence of fall related psychological concern in the sample with pain and control group if present and any correlations between pain and psychological concerns related to falling. Methodological risk assessment We used the Newcastle Ottawa Scale (NOS; [24]) to assess the methodological quality of all included articles. We included studies without a control group due to the paucity of data [23] and such studies were treated as cross-sectional case controlled studies for the purposes of methodological assessment. The inclusion of observational studies without a control group in

Pain and fear of falling in older adults

systematic reviews is justified when there is a paucity of literature and reporting such studies can provide valuable information but the results must be interpreted with caution [23]. Naturally it was anticipated these studies to have a low methodological quality rating. The NOS assesses the quality of non-randomised trials and its validity and reliability has been established [24] and such criterion is particularly essential when critically appraising studies without a control group [23]. The NOS focuses on three main methodological features: (1) the selection of the groups, (2) the comparability of the groups and (3) the ascertainment of the outcome of interest. The NOS has predefined scoring criteria, but some of these can be further specified for the topic of study. We adapted the NOS to take account of age, gender and/or comorbidity as comparability measures. In addition, we adapted the NOS to consider the measurement of psychological concerns related to falls in the exposure category. The maximum score that any study can achieve on the NOS is 9 points. Studies that score 5 or more are normally considered of satisfactory methodological quality [25]. Due to the paucity of literature we included studies with a score less than this (due to the absence of a control group) but present the results with caution [23]. The methodological assessment was independently completed by two authors and consensus was reached through discussion. Summary measures and data analysis

Records idenfied through database searching (n =892)

Study selection The original search yielded 892 articles, which was reduced to 568 after the removal of duplicates. Two additional papers were identified as potentially eligible through additional sources (reference lists of identified articles). The titles, key words and abstracts of each of these articles were screened and when the eligibility criteria were applied, 56 articles were included in the full text review. At the full text review, 44 articles were excluded with reasons (see Figure 1). Common factors for exclusion included an absence of measurement of psychological concern related to falls in the sample (n ¼ 17), no assessment of pain (n ¼ 9) or age (n ¼ 6). After discussion between two reviewers 12 articles were included in the narrative review. Full details of the search strategy are available in Figure 1. Study characteristics A total of 12 studies, with information on 3398 older adults, were included in the review. Details of the studies are summarised in Table 1. Six of these studies employed a cross-sectional design [26–31], three were cohort studies [32–34] and the remaining three studies consisted of a cross-sectional design but the sample consisted only of only older adults with pain (i.e. no control group) [35–37]. Thus, the exact number of individuals with pain and controls within the review is not known, since several of the studies did not provide a breakdown of the sample characteristics for these two groups of interest. Only two studies provided clear and separate data for the sample characteristics for those with and

Addional records idenfied through other sources 2 Reference lists of arcles

Records aer duplicates removed (n =568)

Records screened (n = 568)

Eligibility

Screening

Idenficaon

Meta-analysis was not appropriate as several different falls related psychological constructs were apparent within the included studies.

1933

Results

Full-text arcles assessed for eligibility (n = 56)

Included

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

DOI: 10.3109/09638288.2014.882419

Studies included in narrave synthesis (n = 12)

Records excluded with reasons (n = 512)

Full-text arcles excluded (n=44), with reasons: N=17 – no measure of falls related psychological concern N=6 age N= 9 did not assess pain N=3 Non community sample N=3 language N= 1 spinal cord injury sample N= 1 post fracture sample N=1 dissertaon abstract N=1 contacted author 3 mes for info and no response N=2 concerns over methodological quality

Figure 1. PRISMA (Moher et al. [22]) flow diagram for search strategy.

Cross-sectional cultural validation of FES-I Community (Greece)

Case controlled community (Canada)

Prospective cohort study Community (Aus)

Cross-sectional Community (USA)

Cohort study (Pre and Post falls intervention, data in paper looks at risk factors pre and post falls intervention) Community (Canada)

Cohort study Community (Canada)

Champagne et al. [27]

Cumming et al. [32]

Fessell & Nevitt [35]

Fletcher et al. [33]

Hadjistavropoulos et al. [34]

Design and setting

Billis et al. [26]

Study

Table 1. Summary of included studies.

N ¼ 89 39 (43.8%) female 72.9 ± 6.04 years Convenient sample in 3 local day centres. No separate data on those reporting bodily pain. N ¼ 30 100% female CLBP N ¼ 15 females with 68.9 + 6.6 years Recruited from local community. Control group N ¼ 15 females without CLBP 69.4 + 6.4 years No pain in previous year and never experienced disabling CLBP N ¼ 418 completed baseline FES 77.0 ± 7.0 years 57% female All community dwelling but recruited from hospital (84%), outpatient clinics (5%) and local day centres (11%). No separate data on those reporting body pain. N ¼ 570 all had Rheumatoid arthritis 75.8% female 64.9 ± 8.5 years All sample reported pain at one or more joint Recruited randomly from local rheumatologists. No control group. N ¼ 559 81.0 ± 6.4 years 65.8% (n ¼ 368) female. All recruited from a falls prevention initiative. No separate data on those reporting pain. N ¼ 571 (79 did not finish study) 30% response rate 76.6 ± 5.4 years 64% female Retired local citizens It is unclear exactly how many had pain how many were controls. The mean MPI score was 5.0 at baseline for the sample, therefore no clear control group.

Sample characteristics and recruitment sources

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

B. Stubbs et al. Multidimensional pain inventory pain severity scale [39]. Three items rating pain from 0 to 6, higher scores indicating higher severity of pain. Authors used total of 3 items to measure pain. Authors add up MPI pain severity scores, unclear if all had some degree of pain or not. However mean MPI pain severity at baseline 5.0 ± 4.7.

Pain scale as independent variable from the InterRAI CHA.

Asked about pain at 18 joint sites. Pain severity assessed on scale from 0 to 100.

SF 36 score bodily pain taken at baseline and at 12 month follow-up. Unclear how many reported pain and no pain.

64months CLBP Presented tension, soreness, and/or stiffness in the lower back region with radiating pain limited to the buttocks

SF-36 v2 bodily pain subscale. Pain rated over the past month. Unclear how many reported pain and no pain.

Pain ascertainment

1934 Disabil Rehabil, 2014; 36(23): 1931–1942

Case controlled study Community (Aus)

Cross-sectional Community (Canada)

Cross-sectional Community (UK)

Cross-sectional study Community (USA)

Cross-sectional Community (Japan)

Levinger et al. [28]

Martin et al. [29,37]

Martin et al. [29]

Moore et al. [30]

Suziki et al. [31]

N ¼ 82 100% women 73.8 ± 8.1 years 100% osteoporosis Stratified by NRS scores: Mild pain (0–3) n ¼ 19 73.3 + 7.7years Moderate pain (4–6) n ¼ 51 74.5 + 8.2 years. Severe pain (7–10) n ¼ 12 71.7 ± 8.1 years No control group as such, but mild pain group was used as a reference group. N ¼ 62 OA group (pre and post-test TKR, baseline data used): N ¼ 35, 67 ± 7 years 45% female. All had OA and 100% had knee pain of varying severity. Control group: N ¼ 27, 65 ± 11 years (ns) 53% female (ns) Neither OA nor pain in knees. N total ¼ 65 60–90 years (mean 64.3 ± 2.41) N ¼ 36 55.5% female 70–100 years (mean 77.17 ± 4.88) N ¼ 29 48.2% female All attending physiotherapy clinic and had pain at one or more location, no control group. N ¼ 684 100% female 64.2 ± 6 years Convenient sample from primary care practice practices. No separate data for those with and without pain. N ¼ 133 77.4% female 74.1 ± 9.5 years Recruited from local community. No separate data for those with and without pain. N ¼ 135 92 females (68%) 82.2 years ± 6.83 43 males 76.14 ± 7.37 Day centre Service users No separate data for those with and without pain

SF 36 bodily pain subscale 0–100. Unclear how many reported pain and no pain.

SF 36 bodily pain subscale 0–100 28.4% sample had mod/severe bodily pain in past month 19.6% sample had mod/severe bodily pain interfering with activities. SF 36 short form V2. Unclear how many reported pain and no pain.

Multidimensional pain inventory (pain severity) score, items rated from 0 to 6, higher scores indicating higher levels of pain.

WOMAC. Current pain/severity unknown

NRS for back pain

FES-I, falls efficacy scale International; SF 36, Short form 36; CLBP, chronic low back pain; MPI, multidimensional pain inventory; NRS, numerical rating scale; WOMAC, Western Ontario and McMaster Universities Arthritis Index; OA, osteoarthritis; NS, non-significant; V2, version 2.

Cross-sectional Community (Germany)

Hubscher et al. [36]

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

DOI: 10.3109/09638288.2014.882419

Pain and fear of falling in older adults 1935

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

1936

B. Stubbs et al.

without pain [27,28]. In the case of one study, it was not clear whether they had a control group [34] while six other studies [26,30–33,37] did not provide separate sample characteristics for those with and without pain. None of the included studies had the primary aim of establishing an association between pain and psychological concerns related to falls. The sample size within each study varied, ranging from the smallest with 30 older adults (15 with chronic low back pain (CLBP) and 15 controls; [27]) to the largest which had 684 participants [29]. The mean age ranged from 64.2 years [29] to 82.2 years [31] (in the female sample). There was considerable heterogeneity between the comorbidities observed within the samples of each of the included studies and this is summarised in Table 1. In the two studies with a clear control group [27,28] there were no statistically significant differences observed in the age, gender or comorbidity characteristics of the two groups (those with and without pain). In addition, the participants within each study were recruited through a range of different sources, summary information on this and wider demographic information is presented in Table 1. Table 1 also lists the different methods by which pain was assessed. The most common was the Short Form 36 [38] bodily pain subscale which was used in five studies [26,29–32]. Two studies used a numerical rating scale [35,36] and two studies used the multidimensional pain inventory pain severity scale [34,37,39]. Methodological quality of the included studies The mean NOS score for the included articles was 5.3 ± 1.8 and are summarised in Supplementary Table (Online supplementary information only). The methodological quality of the cohort studies was higher 7.0 ± 0.0 compared to the case controlled studies 4.8 ± 1.8 Four studies scored less than 5 on the NOS raising concerns about the risk of bias and methodological rigour in those studies, which could have influence any observed results reported within their studies. Three of these studies [35–37] had low NOS because they did not have a control group and were treated as a case controlled study and the absence of a control group meant they would naturally score low on the NOS. Each of these studies scored zero (out of a possible 5) that compares the pain and control groups on selection, comparability and exposure.

Disabil Rehabil, 2014; 36(23): 1931–1942

measuring FOF [31,35] and two investigated whether participants restricted their activities due to FOF [29,33]. In addition, two authors used the SAFFE or MSAFFE to assess FOF [30,34]. Consequences of falling Only one study [30] measured CoF using the COF scale [14]. Six [26,29–30,32,34,37] of the included articles discussed the psychometric properties of the instrument they used to measure psychological concerns related to falling and all reported that these were acceptable. However, none specifically discussed the psychometric properties in samples of older adults with pain (see Table 2). Prevalence of psychological concerns related to falling in older adults with pain compared to those without pain Only two studies reported the prevalence of fall related psychological concerns in a sample with and without pain. Champagne et al. [27] established that those with CLBP had significantly lower balance confidence than the control group (ABC score 79.5 ± 17.2 versus 93.5 ± 4.6 in control group p50.005). In the other study, Levinger et al. [28] established those with knee pain had higher concerns about falling (lower falls efficacy) than controls (FES-I 11.4 ± 3.0 versus 7.6 ± 1.2 in control p50.05). The prevalence of psychological concerns related to falling was reported in seven other studies [26,29,30,33,34,36,37] but these studies failed to report information about a clear control group (a group without pain). This means a comparison between individuals with and without pain is not possible. Three studies [31–33] failed to report the prevalence of fall related psychological concerns. These findings are presented in Table 2. Associations between pain and psychological concerns related to falling In total, 10 studies [26,27,29–36] reported an association between pain and psychological concerns related to falling in their published papers and all of these except one [31] established at least one significant association. The associations are summarised in Table 2.

Measurement of fall related psychological concerns

The association between pain and falls efficacy

A diverse range of measures were used and several constructs of falls related psychological concerns were evaluated and are summarised in Table 2. Table 3 provides a list of studies investigating each of the falls related psychological concerns.

Seven studies [26,28,30,32,34,36,37] measured falls efficacy, three [26,30,36] of which used the falls efficacy scale international (FES-I [40]), one [28] used the short FES I [41] and three [32,34,37] used the falls efficacy scale (FES; [42]).

Two authors [26,30] established that FES-I scores were correlated with bodily pain while another established that low baseline FES scores (575) were associated with deterioration in bodily pain scores of 17.7 (p50.05) over 12 months [32]. Hadjistavropoulos et al. [34] established that the FES was a predictor of future falls in their sample who scored a mean of 5.0 on the MPI. Hubscher et al. [36] established a linear relationship between pain and falls efficacy; those with severe pain (7–10 on NRS) had a significantly increased FES-I score (more concerned about falling).

Balance confidence

The association between pain and balance confidence

Four [27,30,34,37] studies measured balance confidence using the activities balance confidence scale (ABC; [11]).

In a sample of CLBP, ABC scores were correlated to the Oswestry disability index scores (ODI) indicating those with higher levels of disability had significantly reduced balance confidence (0.60, p50.05). Hadjistavropoulos et al. [34] found that ABC scores were predictors of future fallers in their sample (their sample scored a mean score of 5.0 on the MPI). Another study found a moderate (r ¼ 0.42, p50.01) correlation between bodily pain and ABC scores [30].

Falls efficacy

Fear of falling and avoidance of activities due to FOF In total, six studies [29–31,30–35] investigated whether there was an association between pain and FOF in their sample. Four studies [29,31,33,35] used a single item question, with two

Psychological concern related to falling construct measured

Falls efficacy

Balance confidence

Falls efficacy

Fear of falling and avoidance of activities due to a fear of falling

Avoidance of activities due to a fear of falling

Fear of falling and activity restriction (SAFFE)

Study

Billis et al. [26]

Champagne et al. [27]

Cumming et al. [32]

Fessell & Nevitt [35]

Fletcher et al. [33]

Hadjistavropoulos et al. [34]

Table 2. Psychological concerns related to falling.

All respondents were asked if they limited going outdoors because fear of falling. (Activity restriction due to FOF dichotomised yes/no). SAFFE includes 3 subscales (FOF, activity avoidance and activity level). Authors report

Asked single item question: If they had any fears of falling, those answering yes asked to classify little, somewhat or very FOF. All subjects asked if they limited activities due to concerns or FOF (Yes/no). Psychometric properties not discussed.

FES to assess falls related self-efficacy. 10 item questionnaire rate self-efficacy on tasks from 0 (low falls self-efficacy) to 10 (high falls self-efficacy) giving total score from 0 to 100. Authors report FES has good internal consistency (a ¼ 0.91), test–retest reliability (r ¼ 0.71) and construct validity [42].

FES-I Greek version validation: Criterion related validity correlation with other measures: CONFbal (r ¼ 0.694, p50.01) Single item FOF question (r ¼ 0.769, p50.01) Test–retest reliability ICC 0.951 (SEM 1.79 and SDD 20.44%) ABC scale Psychometric properties not discussed in article.

Measurement of psychological concerns of falling

Unclear how many had pain, but mean MPI pain severity at baseline

Not available for those with or without pain.

All sample had pain at 14 joint: Level of FOF: None 49.5% A little 24.2% Somewhat 16.7% Very 9.6% Activity avoidance due to FOF: Yes 37.9% No 62.1%

ABC scores CLBP group 79.5 ± 17.2* p50.005 Control 93.5 ± 4.6 Not available for pain and no pain samples.

Not available for pain and no pain samples.

Prevalence of psychological concerns of falling

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

Pain and fear of falling in older adults (continued )

FES predicts future falls in sample* (OR 0.56, CI 0.42 to 0.75, p50.01)

FES at baseline and changes in SF 36 bodily pain scores from baseline to 12 months: Bodily Pain scores changed more than any other QOL measure in the study. Low baseline FES (575) associated with deterioration in pain scores 17.7*  and 19.4 E* both p50.05 Moderate baseline FES scores (76–99) associated with deterioration in pain scores 4.6  and 4.7 E (NS) High baseline FES scores (100) associated with improved pain 2.7 and 3.1 E (NS) FOF group correlates (n ¼ 287) Number of mean painful joints* p50.01 (1.53 versus 0.99 in no FOF) Number of mean painful lower limb joints* p50.01 (11.4 versus 8.6 in no FOF) Pain severity mean* p50.01 (50.1 versus 34.5 in no FOF group). Activity avoidance correlates due to FOF: Number of mean painful joints* p50.01 (11.6 versus 9.0 in no Activity limitation group) Number of mean painful lower limb joints* p50.01 (4.5 versus 3.4 no Activity limitation group) Pain severity mean * p50.01 (51.6 versus 36.7 no Activity limitation group). Logistic regression – pain in lower limbs (OR1.20, CI 1.08 to 1.34, p50.01) significant risk of FOF and in activity limitation due to FOF (OR 1.15, CI 1.03 to 1.28, p50.05). Pain predictor of activity limitation due to FOF OR 1.78 (CI ¼ 1.41 to 2.24*, p50.0001)

ABC scores in CLBP group correlated with: ODI scores 0.60* p50.05

FES-I scores correlated to SF-36 bodily pain subscale 0.363* p50.01

Associations between pain and psychological concerns related to falling

DOI: 10.3109/09638288.2014.882419

1937

Falls efficacy

Falls efficacy

Falls efficacy Balance confidence

Avoidance of activities due to fear of falling

Balance confidence Falls efficacy Fear of falling and activity restriction (mSAFFE) Concerns regarding the consequences of falling scale.

Levinger et al. [28]

Martin et al. [29,37]

Martin et al. [29]

Moore et al. [30]

Balance confidence Falls efficacy

Psychological concern related to falling construct measured

Hubscher et al. [36]

Study

Table 2. Continued

Not reported.

Baseline Short FES-I scores Pain group 11.4 ± 3.0* p50.05 Control group 7.6 ± 1.2 100% sample had pain. 60–69 year olds: FES 90.97 ± 24.69 ABC 76.96 ± 21.52 70–100 year olds: FES 85.55 ± 17.8 ABC 73.6 Not reported for those with and without pain.

Unclear how many had pain and data not available for those with and without pain.

Compared pain severity with FES-I scores Mild versus moderate p ¼ ns Mild versus severe p ¼ 0.029* Moderate versus severe p ¼ 0.069 All groups p ¼ 0.042* Controlling for age, fracture status and history of falls.

FES-I scores Mild pain group: 25.9 ± 7.7 Moderate pain group: 28.3 ± 10.6 Severe pain group: 35.7 ± 8.1

62.9% with FOF had moderate to severe body pain versus 24.4% in non FOF group* p50.01 55.7% with FOF had pain moderate to severely interfering with activities versus 15.5% non FOF group* p50.01 Bivariate correlation: FOF & mod/severe pain: OR 5.36 (3.1 to 9.4) p50.01 * FOF & pain mod/severe interfering with activities OR 6.91 (4.0 to 12.1) p50.01 * Bodily pain associated with: FES-I 0.32* (CI: 0.46 to 0.16) p50.01 ABC 0.42* (CI: 0.27 to 0.55) p50.01 mSAFFE 0.32* (CI: 0.46 to 0.16) p50.01 CoF 0.24*p50.01

Authors conduct a number of analyses but data not available for older adults in the sample.

ABC predicts future falls * (OR 1.04, CI 1.01 to 1.06, p50.01) SAFFE not able to predict falls.

Associations between pain and psychological concerns related to falling

5.0 ± 4.7 and scores were: SAFFE FOF 0.35 ± 0.42 Activity avoidance 3.0 ± 2.85 Activity level 9.0 ± 2.38 ABC 83.0 ± 19.5 FES 9.0 ± 1.54

Prevalence of psychological concerns of falling

B. Stubbs et al.

ABC FES-I mSAFFE modified survey of activities and fear of falling in elderly Authors discuss validity and internal consistency (cronbach’s alpha 0.91–0.92, Yardley and Smith).

Single item question ‘in the past 12 months have you limited your activities because you are afraid you will fall?’ Yes/no (n ¼ 70, 10%). Authors report single item question correlates well with SAFFE and the FES.

FES ABC Author reports both have satisfactory psychometric properties.

SAFFE has good psychometric properties. ABC 16 item measure to assess balance confidence each item rated 0–100 (higher scores indicating higher balance confidence). Authors report alpha coefficients of 0.96 (time 1) and 0.97 (time 2). FES 10 items to address confidence in carrying out activities without falling (falls self-efficacy). Alpha coefficients of 0.96 (time 1) and 0.97 (time 2). FES-I German validated version. Psychometric properties not reported in article. 16 item scale looking at FOF. Answers from not concerned (1) to very concerned (4). Scores from 16 (no concern of FOF) to 64 (severe concern about falling). Kempen et al. [41] ICC 0.79 Internal reliability mean above 0.90 Short FES-I [41]. Psychometric properties not discussed in article. 7 item Likert scale measure of fear of falling across 7 activities.

Measurement of psychological concerns of falling

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

1938 Disabil Rehabil, 2014; 36(23): 1931–1942

Pain and fear of falling in older adults

FES-I, falls efficacy scale international; FOF, fear of falling; ICC, interclass correlation coefficient; SEM, standard error of the mean; SF 36, short form 36; ABC, activities balance confidence; CLBP, chronic low back pain; ODI, Oswestry disability Index; FES, falls efficacy scale; QOL, quality of life; NS, non-significant; SAFFE, Survey of Activities and Fear of Falling in the Elderly; OR, odds ratio; mSAFFE, modified Survey of Activities and Fear of Falling in the Elderly; CoF, consequences of falling. *Statistically significant.

Not available for pain and no pain samples Suziki et al. [31]

Fear of falling

CoF Consequences of falling scale Reliability confirmed through internal consistency estimates ranging 0.86 to 0.94. Single item question: ‘at the present time, are you very fearful, somewhat fearful or not fearful that you may fall?’ Afken et al., 1994. Psychometric properties not discussed.

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

No associations between moderate FOF or nor very FOF and body pain.

DOI: 10.3109/09638288.2014.882419

1939

The association of pain with FOF and avoidance of activities due to FOF Fessell and Nevitt [35] found a number of relationships between pain and FOF. Those with FOF had a higher mean number of painful joints, higher mean number of painful lower limb joints and higher pain severity mean score compared to those without FOF. The authors found a similar relationship with their sample that was classified as avoiding activities due to FOF. In a logistic regression, the authors established that lower limb pain was associated with significantly increased risk of FOF (OR (odds ratio) 1.20, Confidence Interval (CI) 95%: 1.08–1.34, p50.01) and activity avoidance due to FOF (OR 1.15, CI: 1.03–1.28, p50.05). Fletcher et al. [33] that pain was associated with outdoor activity limitation due to FOF (OR 1.78, CI: 1.41–2.24, p50.0001). Martin et al. [29] established that many more of those with FOF had moderate/severe pain (62.9% versus 24.4%) and moderate/severe pain interfering with activities (55.7% versus 15.5%) compared to the non FOF group (both p50.01). Moore et al. [30] established mSAFFE scores were significantly correlated to bodily pain while Hadjistavropoulos et al. [34] found that unlike the FES and ABC, the SAFFE was unable to predict future falls in their sample. The association of pain and consequences of falling Only one study investigated CoF and found that this was negatively associated with bodily pain [30] (r ¼ 0.24, p50.01).

Discussion To the author’s knowledge, this is the first systematic review to investigate the relationship between pain and psychological concerns related to falling in community dwelling older adults. Despite pain and psychological concerns related to falling being two common and pervasive issues for this population, we did not find one study whose primary objective was to explore this relationship. There was a considerable amount of heterogeneity in the assessment of pain and psychological concerns related to falls, which presented difficulties when interpreting the results. In addition, we encountered heterogeneity in the comorbidities seen in the respective samples between individual studies. However, there is evidence from two studies of good methodological quality that older adults with pain have a reduced balance confidence [27] and lower falls efficacy [28] compared to those without pain. In addition, ten out of the 12 studies reported an association between pain and one of the psychological concerns related to falling; nine of these were significant. Despite the methodological limitations, the evidence seems to suggest that pain is associated with lower falls efficacy, increased FOF, reduced balance confidence and avoidance of activities due to FOF. Only one study investigated whether there is an association between pain and CoF and despite a significant correlation being found, no final conclusions can be made. From our results, there are indications older adults with pain have a reduced falls efficacy with two prospective studies finding particularly interesting results. First, Cumming et al. [32] established a substantial (17.7 points on 0–100 NRS) reduction in bodily pain over 12 months in those with low FES scores (575). Secondly, Hadjistavropoulos et al. [34] established that the FES score was a predictor of future falls in their sample who had a mean score of 5.0 on the MPI. It is not surprising that older adults with pain have reduced falls efficacy, since pain is known to cause gait disturbances [19] and this is likely to decrease an individual’s falls efficacy. The finding from Hubscher et al. [36] indicates that the relationship between pain and falls efficacy could be linear; this would be consistent with previous research that has

1940

B. Stubbs et al.

Disabil Rehabil, 2014; 36(23): 1931–1942

Table 3. List of studies investigating each psychological concern related to falling construct. Psychological concern relate to falling construct Falls efficacy

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

Balance confidence Fear of falling (FOF)/avoidance of activities due to FOF Concerns about the consequences of falling

Studies investigating this construct Billis et al., 2011, Cumming et al., 2000, Hadjistavropoulos et al., 2007, Hubscher et al., 2010, Levinger et al., 2011, Martin et al., 2005, Moore et al., 2011 Champagne et al., 2012, Hadjistavropoulos et al., 2007, Martin et al., 2005, Moore et al., 2011 Fessell & Nevitt 1997, Fletcher et al., 2010, Hadjistavropoulos et al., 2007, Martin et al., 2006, Moore et al., 2011, Suziki et al., 2002 Moore et al., 2011

demonstrated that increasing pain intensity is associated with greater physical impairment and increasing falls risk [44,45]. Falls efficacy would be expected to decrease alongside these known changes. The changes in falls efficacy may be an important contributing factor in this previously observed relationship. Our results provide some indication that pain may be associated with reduced balance confidence in community dwelling older adults, although no strong conclusions can be drawn from our results. One study [27] demonstrated older adults with CLBP had significantly lower ABC scores compared to the age matched controls with a mean difference of 14% (p50.005) and demonstrated a large association with ODI (0.60, p50.05). Pain is closely linked to disability in older adults [46] and affected by other factors such as depression and anxiety [8] and it is not surprising that individuals with chronic pain in particular may have reduced confidence in completing their ADL without falling over. However, the sample size in Champagne et al. [27] was very small (n ¼ 30) and although interesting the results are clearly not generalisable. Hadjistavropoulos et al. [34] established that baseline ABC scores could predict future falls, but the association (OR 1.04, CI: 1.01–1.06, p50.01) was small. In another study, Moore and colleagues [30] utilised four measures to capture each of the psychological concerns related to falling and found that the strongest correlation was between ABC scores and bodily pain. Interestingly, the authors also established that the ABC was able to accurately predict an older adult’s propensity to fall. The results from this review provide reasonable evidence that pain is associated with FOF and avoidance of activities due to a FOF. Two studies with large sample sizes [29,35] both found that those classified as having FOF and activity avoidance due to FOF, were likely to report having pain across more body sites and also that the pain they experienced was of greater intensity. Martin et al. [29] reported an odds ratio of 5.36 (CI: 3.1–9.4) and 6.91 (CI: 4.0–12.1) to quantify the relationship with moderate to very severe body pain and pain interfering with activities, respectively. The NOS score for this study was high (6) meaning that the methodological quality was acceptable and the risk of bias is likely to be lower than in other studies. Although the Fessell and Nevitt [35] provided interesting results, it should be noted this study gained a low score on the NOS scale due to the lack of a control group to enable comparison and caution should be asserted with conclusions from this study. Interestingly, although Hadjistavropoulos et al. [34] established falls efficacy and balance confidence were able to predict future falls, the authors established that the SAFFE was not able to predict future falls. Activity avoidance due to pain is known to occur [16] and it seems plausible that pain may be an important factor that mediates this relationship. A recent meta-analysis [20] established that older adults with chronic musculoskeletal pain are less active than those without pain, and activity avoidance due to a FOF could possibly contribute to this observed reduction in physical activity and warrants investigation.

Clinical implications Pain is frequently encountered by community dwelling older people and the FOF and other psychological concerns related to falls is the most common anxiety reported in older people above others such as fear of being robbed and attacked [14,47]. There is increasing evidence that pain is an important factor associated with lower levels of physical activity [20] and actual falls [21,43,44]. The presence of any of the psychological concerns related to falls may be an important factor influencing older adults with pain levels of physical activity and falls. Clinicians working with the older adult, who present with pain and in particular chronic pain, should consider using a detailed falls assessment, including attention to the presence of any psychological concerns related to falling. There is some confusion and overlap between each of the constructs within the spectrum of psychological concerns related to falling. Clinicians should therefore ensure that they are employing the correct outcome measure to capture the phenomenon they wish to measure. In order to avoid confusion, other researchers have suggested that clinicians should simply refer to the terms falls related psychological concerns [8]. Limitations There are a number of considerations when interpreting the results of this review. First, none of the included studies set out to explore a relationship between pain and psychological concerns related to falls, nor did they set out to establish if this is more prevalent in older adults with pain. Although each of the studies in this review produced useful information about the relationship of interest it was not the focus of any one of the studies; this makes it much more difficult to examine the relationship in a systematic review. Second, there was great variation in the methods used to assess pain in the included studies. Encountering such heterogeneity makes it very difficult to quantify the exact location, nature (e.g. musculoskeletal) and severity of pain and makes comparison across studies difficult. Future work should seek to address this. In addition, we encountered a great deal of heterogeneity in the outcome measures used to assess the psychological concerns related to falling, rendering it impossible to conduct a meta-analysis. Further, there was some confusion and mismatching in studies with some reporting they were measuring FOF when they had, in fact, used a falls efficacy instrument (i.e. measuring falls efficacy and not FOF). This has been reported in other reviews [5,6] and it is important that future research clearly defines the central construct or constructs and that they employ the appropriate tool to capture that construct or constructs. In addition, there are some methodological concerns about the studies included in this review as indicated by the low NOS scores in some cases. Four studies were found to have a NOS score below 5. In three cases this was attributable due to the absence of a control group and it is important that caution is attached to any conclusions drawn from these studies. Despite these issues, we felt these studies warranted inclusion anyway due to the paucity of literature in the area and because they did

DOI: 10.3109/09638288.2014.882419

find significant relationships among the variables of interest, albeit that may be attributed partially due to a lack of methodological rigour. In addition, there was considerable heterogeneity observed between the sample populations within each study of the included studies. Lastly, most (9/12) of these studies were crosssectional making it not possible to infer causality and explore the exact nature of the association between pain and concerns about falling. Future prospective work would be beneficial to untangle the undoubtedly complex, possibly recursive, relationship between pain and psychological concerns related to falls.

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

Future research There is a need for well-designed research to establish the relationship between older adults with pain and each of the psychological concerns related to falling. Future work should clearly assess the location, type, nature and severity of pain in the sample and define and accurately measure one or more psychological concerns related to falling. In addition, it is important that future works utilises a homogenous comparison group so that stronger conclusions can be drawn from any results. Future work should also explore the relationship between the different psychological concerns related to falls and physical activity and actual falls. This work should inform future interventional studies to address these issues in older adults with pain and be applicable to clinicians in practice.

Conclusion This review provides new insight that older adults with pain may be more susceptible to experiencing psychological concerns related to falls, in particular FOF, activity avoidance due to FOF and reduced falls self-efficacy. Pain is a common and pervasive problem in older adults and the association with these factors is likely to impair such individuals functioning and wellbeing further and may be an important moderator in the lower levels of physical activity and increased risk of falls seen in this group. Future research is required to establish which of the falls related psychological constructs are particularly problematical in older adults with pain so that future interventional work can be developed and inform clinical practice.

Acknowledgements We thank Dr Andy Soundy for providing feedback on a draft of the article.

Declaration of interest B.S. is supported from a Vice Chancellors scholarship at the University of Greenwich. This did not affect the research at any stage or the decision to publish. None to declare from the other authors.

References 1. Gillespie LD, Robertson MC, Gillespie WJ, et al. Interventions for preventing falls in older people living in the community. Cochrane Database Syst Rev (Online) 2012;9:CD007146. 2. Deandrea S, Bravi F, Turati F, et al. Risk factors for falls in older people in nursing homes and hospitals. a systematic review and meta-analysis. Arch Gerontol Geriatr 2013;56:407–15. 3. Deandrea S, Lucenteforte E, Bravi F, et al. Risk factors for falls in community-dwelling older people: a systematic review and metaanalysis. Epidemiology 2010;21:658–68. 4. World Health Organisation. WHO global report on falls prevention worldwide. Available from http://www.who.int/ageing/publications/ Falls_prevention7March.pdf [last accessed 16 Jul 2013].

Pain and fear of falling in older adults

1941

5. Jørstad EC, Hauer K, Becker C, Lamb SE; ProFaNE Group. Measuring the psychological outcomes of falling: a systematic review. J Am Geriatr Soc 2005;53:501–10. 6. Moore DS, Ellis R. Measurement of fall-related psychological constructs among independent-living older adults: a review of the research literature. Aging Ment Health 2008;12:684–99. 7. Scheffer AC, Schuurmans MJ, van Dijk N, et al. Fear of falling: measurement strategy, prevalence, risk factors and consequences among older persons. Age Ageing 2008;37:19–24. 8. Hull SL, Kneebone II, Farquharson L. Anxiety, depression, and fallrelated psychological concerns in community-dwelling older people. Am J Geriatr Psychiatr 2013;21:1287–91. 9. Tinetti ME, Powell L. Fear of falling and low self-efficacy: a cause of dependence in elderly persons. J Gerontol 1993;48:35–8. 10. Bandura A. Self-efficacy: toward a unifying theory of behavioral change. Psychol Rev 1977;84:191–215. 11. Powell LE, Myers AM. The activities-specific balance confidence (ABC) scale. J Gerontol: Med Sci 1995;50A:M28–34. 12. Delbaere K, Crombez G, Vanderstraeten G, et al. Fear-related avoidance of activities, falls and physical frailty. A prospective community based cohort study. Age Ageing 2004;33:368–73. 13. Zijlstra GA, van Haastregt JC, van Eijk JT, et al. Prevalence and correlates of fear of falling, and associated avoidance of activity in the general population of community-living older people. Age Ageing 2007;36:304–9. 14. Yardley L, Smith H. A prospective study of the relationship between feared consequences of falling and avoidance of activity in community-living older people. Gerontologist 2002;42:17–23. 15. Sions JM, Hicks GE. Fear-avoidance beliefs are associated with disability in older American adults with low back pain. Phys Ther 2011;91:525–34. 16. Camacho-Soto A, Sowa GA, Perera S, Weiner DK. Fear avoidance beliefs predict disability in older adults with chronic low back pain. PM R 2012;4:493–7. 17. Vincent HK, Seay AN, Montero C, et al. Kinesiophobia and fearavoidance beliefs in overweight older adults with chronic low-back pain: relationship to walking endurance–part II. Am J Phys Med Rehabil 2013;92:439–45. 18. Abdulla A, Adams N, Bone M, et al. Guidance on the management of pain in older people. Age Ageing 2013;42:i1–57. 19. Leveille SG, Bean J, Bandeen-Roche K, et al. Musculoskeletal pain and risk for falls in older disabled women living in the community. J Am Geriatr Soc 2002;50:671–8. 20. Stubbs B, Binnekade TT, Soundy A, et al. Are older adults with chronic musculoskeletal pain less active than older adults without chronic pain? A systematic review and meta-analysis. Pain Med 2013;14:1316–31. 21. Stubbs B, Binnekade T, Eggermont L, et al. Pain and the risk for falls in community-dwelling older adults: a systematic review and meta-analysis. Arch Phys Med Rehabil 2014;95:175–87. 22. Moher D, Liberati A, Tetzlaff J, Altman DG, The PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. 23. Fitzpatrick-Lewis D, Ciliska D, Thomas H. The methods for the synthesis of studies without control groups. Hamilton (ON): National collaborating centre for methods and tools, Available from: http://www.nccmt.ca/pubs/non-RCT2_EN.pdf [last accessed 23 July 2013]. 24. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in metaanalyses. Available from: http://www.ohri.ca/programs/clinical_ epidemiology/oxford.asp [last accessed 25 Jan 2013]. 25. Kwon BK, Roffey DM, Bishop PB, et al. Systematic review: occupational physical activity and low back pain. In Occup Med (Lond) 2011;61:541–8. 26. Billis E, Strimpakos N, Kapreli E, et al. Cross-cultural validation of the Falls Efficacy Scale International (FES-I) in Greek community-dwelling older adults. Disabil Rehabil 2011;33: 1776–84. 27. Champagne A, Prince F, Bouffard V, Lafond D. Balance, fallsrelated self-efficacy, and psychological factors amongst older women with chronic low back pain: a preliminary case-control study. Rehabil Res Pract 2012;2012:article ID 430374. 28. Levinger P, Menz HB, Wee E, et al. Physiological risk factors for falls in people with knee osteoarthritis before and early after knee

1942

29.

30.

31.

32.

33.

Disabil Rehabil Downloaded from informahealthcare.com by Nyu Medical Center on 05/24/15 For personal use only.

34.

35.

36.

B. Stubbs et al.

replacement surgery. Knee Surg Sports Traumatol Arthrosc 2011;19: 1082–9. Martin FC, Hart D, Spector T, et al. Fear of falling limiting activity in young-old women is associated with reduced functional mobility rather than psychological factors. Age Ageing 2005;34: 281–7. Moore DS, Ellis R, Kosma M, et al. Comparison of the validity of four fall-related psychological measures in a community-based falls risk screening. Res Q Exerc Sport 2011;82:545–54. Suziki M, Ohyama N, Yamada K, Kanamori M. The relationship between fear of falling, activities of daily living and quality of life among elderly individuals. Nurs Health Sci 2002;4:155–61. Cumming RG, Salkeld G, Thomas M, Szonyi G. Prospective study of the impact of fear of falling on activities of daily living, SF-36 scores, and nursing home admission. J Gerontol A Biol Sci Med Sci 2000;55:M299–305. Fletcher PC, Guthrie DM, Berg K, Hirdes JP. Risk factors for restriction in activity associated with fear of falling among seniors within the community. J Patient Saf 2010;6:187–91. Hadjistavropoulos T, Martin RR, Sharpe D, et al. A longitudinal investigation of fear of falling, fear of pain, and activity avoidance in community-dwelling older adults. J Aging Health 2007;19: 965–84. Fessel KD, Nevitt MC. Correlates of fear of falling and activity limitation among persons with rheumatoid arthritis. Arthritis Care Res 1997;10:222–8. Hu¨bscher M, Vogt L, Schmidt K, et al. Perceived pain, fear of falling and physical function in women with osteoporosis. Gait Posture 2010;32:383–5.

Disabil Rehabil, 2014; 36(23): 1931–1942

37. Martin RR, Hadjistavropoulos T, McCreary DR. Fear of pain and fear of falling among younger and older adults with musculoskeletal pain conditions. Pain Res Manag 2005;10:211–19. 38. Ware JE. SF 36 health survey: manual and interpretations guide. Boston (MA): Health institute; 1993. 39. Kerns RD, Turk DC, Rudy TE. The west Haven Yale Multidimensional Pain Inventory (WHYMPI). Pain 1985;23: 345–56. 40. Yardley L, Beyer N, Hauer K, et al. Development and initial validation of the Falls Efficacy Scale-International (FES-I). Age Ageing 2005;34:614–9. 41. Kempen GI, Yardley L, van Haastregt JC, et al. The Short FES-I: a shortened version of the falls efficacy scale-international to assess fear of falling. Age Ageing 2008;37:45–50. 42. Tinetti ME, Richman D, Powell L. Falls efficacy as a measure of fear of falling. J Gerontol: Psychol Sci 1990;45:P239–43. 43. Leveille SG, Bean J, Bandeen-Roche K, et al. Musculoskeletal pain and risk for falls in older disabled women living in the community. J Am Geriatr Soc 2002;50:671–8. 44. Leveille SG, Jones RN, Kiely DK, et al. Chronic musculoskeletal pain and the occurrence of falls in an older population. JAMA, J Am Med Assoc 2009;302:2214–21. 45. Sturnieks DL, Tiedemann A, Chapman K, et al. Physiological risk factors for falls in older people with lower limb arthritis. J Rheumatol 2004;31:2272–9. 46. Scheele J, Enthoven WT, Bierma-Zeinstra SM, et al. Characteristics of older patients with back pain in general practice: BACE cohort study. Eur J Pain 2014;18:279–87. 47. Howland J, Peterson EW, Levin WC, et al. Fear of falling among the community-dwelling elderly. J Aging Health 1993;5: 229–43.